Our overall goal is to identify how breast cancer cells modulate transcription in surrounding bone marrow stroma to engender a supportive environment. Despite the importance of cancer cell-stromal interactions, current approaches are poorly suited to analyze the effect of cancer cells on their microenvironment (and vice versa) while within tumors. Our specific project uses a novel cell-specific metabolic labeling strategy to identify stromal genes that are necessary for cancer growth in vivo. We hypothesize that bone metastatic breast cancer growth requires reprogramming of specific bone marrow stromal transcripts that can be discovered through stromal-specific metabolic mRNA labeling, functionally validated in the intact tumor microenvironment and confirmed in clinical bone metastatic breast cancer samples. Specific Aims of this project are to: 1. Measure reprogramming of active transcription in human bone marrow stromal cells and in marrow-derived mesenchymal stem cells (MSCs) resulting from direct contact with breast cancer cells. The approach exploits a protozoan enzyme (UPRT) to enable cell-type specific transcript labeling without requiring microdissection or cell separation. Global reprogramming of stromal cells during direct contact with highly bone-metastatic and less metastatic cancer cells will be measured. 2. Generate stromal cell variants that are resistant to cancer-induced or -repressed transcription of key stromal pathways and test their effect on cancer cell survival, growth and motility in vitro. Suppression of the marrow stromal cell response to cancer cells is predicted to inhibit the growth or motility of breast cancer cells in vitro. We will target genes including cancer-induced stromal NF-?B and Stat1 and cancer-repressed stromal Wnt5a, which our preliminary data implicate as candidate mediators of tumorigenic crosstalk. 3. Determine whether reprogramming of select stromal genes are necessary and/or sufficient for stroma to promote carcinogenesis in vivo, and measure in vivo changes in cell-specific transcription in different tumor microenvironments. This aim tests the hypothesis that reactive changes in select bone marrow stromal genes are necessary for stroma to promote carcinogenesis in vivo. Orthotopic and intraosseous xenograft models combining cancer cells and manipulated stroma are used. 4. Validate stromal reprogramming and stromal effects on cancer in clinical samples. Stromal genes induced (or repressed) in Aims 1-3 are predicted to be induced (or repressed) in clinical samples of bone metastatic breast cancer. Completion of these aims will for the first time identify and functionally validate cancer induced changes in bone marrow stromal transcription as it occurs in vivo; conversely, this will also be the first analysis of stromal-induced changes in breast cancer cell transcription as it occurs in vivo. While this proposal focuses on bone metastases, these methodologies may be applied to other metastatic sites. Our work sets the stage for profiling of specific cells in the tumor environment as they respond to, or resist, pharmacotherapy, radiotherapy, or immunotherapy.